Method for decoding image and apparatus using same

An apparatus for decoding a video supporting a plurality of layers, the apparatus comprising:

• a decoder configured to receive information on a reference layer used for decoding a current picture for interlayer prediction, derive a number of active reference layer pictures used for decoding the current picture based on the information on the reference layer, perform interlayer prediction based on the number of active reference layer pictures to generate a prediction block for a current block, generate a residual block for the current block, and reconstruct the current block based on the prediction block and the residual block,
• wherein the generating the residual block comprises entropy-decoding a bitstream to generate a quantized transformed coefficient and inverse-quantizing the quantized transformed coefficient to generate a transformed coefficient,
• wherein when a layer identifier of a current layer comprising the current picture is not 0 and when a number of reference layer pictures available for interlayer prediction in the same access unit as that of the current picture is not 0, and when a reference layer picture available for interlayer prediction specified by values of maximum temporal sub-layer information on each layer and information on maximum allowed value of temporal sub-layer allowing inter-layer prediction in each layer among direct reference layer pictures comprised in all the direct reference layers of the current layer is present in the same access unit as that of the current picture and included in an interlayer reference picture set of the current picture, the number of active reference layer pictures is derived to be equal to the number of reference layer pictures, and
• wherein the number of reference layer pictures is derived based on information indicating a number of direct reference layers of the current layer, the maximum temporal sub-layer information on each layer, the information on maximum allowed value of temporal sub-layer allowing inter-layer prediction in each layer and a temporal identifier of the current picture, and among the pictures in the direct reference layers of the current layer comprising the current picture, a picture in a reference layer is considered as a reference layer picture available for decoding the current picture for interlayer prediction when maximum temporal sub-layer information of the reference layer is greater than or equal to the temporal identifier of the current picture and either when information on maximum allowed value of temporal sub-layer allowing inter-layer prediction in the reference layer for the current layer is greater than the temporal identifier of the current picture or when the temporal identifier of the current picture is 0.

A method of encoding a video supporting a plurality of layers performed by an encoding apparatus, the method comprising:

• determining information on a reference layer used for encoding a current picture for interlayer prediction;
• deriving a number of active reference layer pictures used for encoding the current picture based on the information on the reference layer;
• performing interlayer prediction based on the number of active reference layer pictures to generate a prediction block for a current block;
• generating a residual block for the current block based on the prediction block;
• and
• encoding the residual block,
• wherein the encoding the residual block comprises transforming the residual block to generate a transformed coefficient and quantizing the transformed coefficient to generate a quantized transformed coefficient,
• wherein when a layer identifier of a current layer comprising the current picture is not 0 and when a number of reference layer pictures available for interlayer prediction in the same access unit as that of the current picture is not 0, and when a reference layer picture available for interlayer prediction specified by values of maximum temporal sub-layer information on each layer and information on maximum allowed value of temporal sub-layer allowing inter-layer prediction in each layer among direct reference layer pictures comprised in all the direct reference layers of the current layer is present in the same access unit as that of the current picture and included in an interlayer reference picture set of the current picture, the number of active reference layer pictures is derived to be equal to the number of reference layer pictures, and
• wherein the number of reference layer pictures is derived based on information indicating a number of direct reference layers of the current layer, the maximum temporal sub-layer information on each layer, the information on maximum allowed value of temporal sub-layer allowing inter-layer prediction in each layer and a temporal identifier of the current picture, and among the pictures in the direct reference layers of the current layer comprising the current picture, a picture in a reference layer is considered as a reference layer picture available for decoding the current picture for interlayer prediction when maximum temporal sub-layer information of the reference layer is greater than or equal to the temporal identifier of the current picture and either when information on maximum allowed value of temporal sub-layer allowing inter-layer prediction in the reference layer for the current layer is greater than the temporal identifier of the current picture or when the temporal identifier of the current picture is 0.

An apparatus for encoding a video supporting a plurality of layers, the apparatus comprising:

• an encoder configured to determine information on a reference layer used for encoding a current picture for interlayer prediction, derive a number of active reference layer pictures used for encoding the current picture based on the information on the reference layer, perform interlayer prediction based on the number of active reference layer pictures to generate a prediction block for a current block, generate a residual block for the current block based on the prediction block, and encode the residual block,
• wherein the encoding the residual block comprises transforming the residual block to generate a transformed coefficient and quantizing the transformed coefficient to generate a quantized transformed coefficient,
• wherein when a layer identifier of a current layer comprising the current picture is not 0 and when a number of reference layer pictures available for interlayer prediction in the same access unit as that of the current picture is not 0, and when a reference layer picture available for interlayer prediction specified by values of maximum temporal sub-layer information on each layer and information on maximum allowed value of temporal sub-layer allowing inter-layer prediction in each layer among direct reference layer pictures comprised in all the direct reference layers of the current layer is present in the same access unit as that of the current picture and included in an interlayer reference picture set of the current picture, the number of active reference layer pictures is derived to be equal to the number of reference layer pictures, and
• wherein the number of reference layer pictures is derived based on information indicating a number of direct reference layers of the current layer, the maximum temporal sub-layer information on each layer, the information on maximum allowed value of temporal sub-layer allowing inter-layer prediction in each layer and a temporal identifier of the current picture, and among the pictures in the direct reference layers of the current layer comprising the current picture, a picture in a reference layer is considered as a reference layer picture available for decoding the current picture for interlayer prediction when maximum temporal sub-layer information of the reference layer is greater than or equal to the temporal identifier of the current picture and either when information on maximum allowed value of temporal sub-layer allowing inter-layer prediction in the reference layer for the current layer is greater than the temporal identifier of the current picture or when the temporal identifier of the current picture is 0.

A non-transitory computer-readable medium storing a bitstream that is generated by a method of encoding a video supporting a plurality of layers, the method comprising:

• determining information on a reference layer used for encoding a current picture for interlayer prediction;
• deriving a number of active reference layer pictures used for encoding the current picture based on the information on the reference layer;
• performing interlayer prediction based on the number of active reference layer pictures to generate a prediction block for a current block;
• generating a residual block for the current block based on the prediction block;
• and
• encoding the residual block,
• wherein the encoding the residual block comprises transforming the residual block to generate a transformed coefficient and quantizing the transformed coefficient to generate a quantized transformed coefficient,
• wherein when a layer identifier of a current layer comprising the current picture is not 0 and when a number of reference layer pictures available for interlayer prediction in the same access unit as that of the current picture is not 0, and when a reference layer picture available for interlayer prediction specified by values of maximum temporal sub-layer information on each layer and information on maximum allowed value of temporal sub-layer allowing inter-layer prediction in each layer among direct reference layer pictures comprised in all the direct reference layers of the current layer is present in the same access unit as that of the current picture and included in an interlayer reference picture set of the current picture, the number of active reference layer pictures is derived to be equal to the number of reference layer pictures, and
• wherein the number of reference layer pictures is derived based on information indicating a number of direct reference layers of the current layer, the maximum temporal sub-layer information on each layer, the information on maximum allowed value of temporal sub-layer allowing inter-layer prediction in each layer and a temporal identifier of the current picture, and among the pictures in the direct reference layers of the current layer comprising the current picture, a picture in a reference layer is considered as a reference layer picture available for decoding the current picture for interlayer prediction when maximum temporal sub-layer information of the reference layer is greater than or equal to the temporal identifier of the current picture and either when information on maximum allowed value of temporal sub-layer allowing inter-layer prediction in the reference layer for the current layer is greater than the temporal identifier of the current picture or when the temporal identifier of the current picture is 0.